fudge factor

Look up fudge factor in Wiktionary, the free dictionary.

Fudge factors are invented variables whose purpose is to force a calculated result to give a better match to what happens in the real world.

Fudge factors are usually calculated retrospectively, and when a calculation has only managed to produce a good match to real data by the addition of a variable that seems artificially tailored to allow that result (to allow the calculations to be "fudged" to give the right answer), critics may sometimes refer to the variable in an uncomplimentary way as a "fudge factor".

The term fudge, meaning falsify, derives from this concept^[1].

Examples in science

Some variables in scientific theory are set arbitrarily according to measured results rather than by calculation (for example, Planck's constant). However, in the case of these fundamental constants, their arbitrariness is usually explicit. To suggest that other calculations may include a "fudge factor" may suggest that the calculation has been somehow tampered with to make results give a misleadingly good match to experimental data.

Cosmological constant

In theoretical physics, when Einstein originally tried to produce a general theory of relativity, he found that the theory seemed to predict the gravitational collapse of the universe: it seemed that the universe should either be expanding or collapsing, and to produce a model in which the universe was static and stable (which seemed to Einstein at the time to be the "proper" result), he introduced an expansionist variable (called the Cosmological Constant) whose sole purpose was to cancel out the cumulative effects of gravitation. He later called this, "the biggest blunder of my life." [1]

Dark matter

Currently there is some controversy over the disagreement between general relativity's predictions and the available astronomical data: In some situations, gravitational effects seem to be acting more powerfully than GR predicts. The current mainstream explanation is that the universe contains a certain amount of unseen dark matter of unknown composition. Due to the vagueness of this explanation, it is not yet clear whether the "dark matter" explanation represents a real discovery, or whether it is an arbitrary "fudge factor" invented to explain away the discrepancy between theory and experiment.

Expected error margins

A common feature of "fudge factors" in science is their arbitrariness, and their retrospective nature.

However, in project management it's common to build a certain error margin into the predicted "resource cost" of a project to make predictions more realistic: there are many unforeseen factors that may delay a project or make it more costly, but very few factors that could result in it being delivered before time or under the calculated budget ... so to some degree, "unexpected" overruns are to be expected, even if their precise nature can't be predicted in advance. Experienced planners may know that a certain type of project will tend to overrun by a certain percentage of its calculated resource requirements, and may multiply the "ideal" calculations by a safety margin to produce a more realistic estimate, and this margin may sometimes be referred to as a fudge factor. However, when planning ahead for expected unpredictabilities, these "error margins" are usually assigned other, more specific names : for instance in warehouse stock control, where a certain amount of stock is expected to disappear naturally through damage, pilfering or other unexplained problems, the discrepancy is referred to as shrinkage.

References

1. ^ fudging - Definitions from Dictionary.com